Skip to contentSkip to conversation

Kernel BUG at mm/slab.c:2380 on NFS server in nfsd (2.6.21.1)

DISCUSSION Roger Heflin 2 participants 3 messages May 16, 2007

Subsystems (2)
NFS client NFS server
lore
Share & tools
lore

Roger Heflin

author
#0 lore raw

Running bonnie over nfs on a RHEL4.4 client against a 2.6.21.1 server got me this crash after about 4 hours of running on the server:

This was running lvm -> ext3 -> nfs ---- nfsclient (RHEL4.4).

Ideas?

Roger

May 15 21:10:31 vault1 kernel: ------------[ cut here ]------------
May 15 21:10:31 vault1 kernel: kernel BUG at mm/slab.c:2380!
May 15 21:10:31 vault1 kernel: invalid opcode: 0000 [1] SMP
May 15 21:10:31 vault1 kernel: CPU 0
May 15 21:10:31 vault1 kernel: Modules linked in: qla2xxx nfsd exportfs lockd nfs_acl sunrpc hidp l2cap bluetooth ipv6 cpufreq_ondemand jfs dm_mirror dm_multipath dm_mod video sbs i2c_ec dock button battery asus_acpi ac lp snd_intel8x0 snd_ac97_codec ac97_bus snd_seq_dummy snd_seq_oss snd_seq_midi_event snd_seq snd_seq_device sg snd_pcm_oss floppy snd_mixer_oss snd_pcm cfi_cmdset_0002 cfi_util mtdpart snd_timer jedec_probe cfi_probe gen_probe snd ck804xrom sata_nv mtdcore chipreg i2c_nforce2 soundcore map_funcs libata snd_page_alloc pcspkr i2c_core k8temp hwmon forcedeth ohci1394 ieee1394 parport_pc ide_cd parport cdrom serio_raw scsi_transport_fc shpchp megaraid_mbox sd_mod scsi_mod megaraid_mm ext3 jbd ehci_hcd ohci_hcd uhci_hcd
May 15 21:10:31 vault1 kernel: Pid: 4256, comm: nfsd Not tainted 2.6.21.1 #1
May 15 21:10:31 vault1 kernel: RIP: 0010:[<ffffffff8025a710>] [<ffffffff8025a710>] cache_alloc_refill+0xe6/0x1f3
May 15 21:10:31 vault1 kernel: RSP: 0018:ffff81021dead6d0  EFLAGS: 00010002
May 15 21:10:31 vault1 kernel: RAX: 0000000000000001 RBX: ffff81012211d960 RCX: ffff810120013040
May 15 21:10:31 vault1 kernel: RDX: 000000000000000e RSI: ffff810139020000 RDI: ffff810120013040
#1 lore raw
Quoted 4 lines · Roger Heflin

Running bonnie over nfs on a RHEL4.4 client against a 2.6.21.1 server got me this crash after about 4 hours of running on the server:

This was running lvm -> ext3 -> nfs ---- nfsclient (RHEL4.4).

Yipes. Has this happened only once, or do you have a way to reliably reproduce it?

Is it a new problem? (And, if so, what changed?)

Quoted 2 lines
May 15 21:10:31 vault1 kernel: ------------[ cut here ]------------
May 15 21:10:31 vault1 kernel: kernel BUG at mm/slab.c:2380!

That's the check_spinlock_acquired() in cache_alloc_refill(). What causes that to fail?

Quoted 13 lines
May 15 21:10:31 vault1 kernel: invalid opcode: 0000 [1] SMP
May 15 21:10:31 vault1 kernel: CPU 0
May 15 21:10:31 vault1 kernel: Modules linked in: qla2xxx nfsd exportfs

lockd nfs_acl sunrpc hidp l2cap bluetooth ipv6 cpufreq_ondemand jfs dm_mirror dm_multipath dm_mod video sbs i2c_ec dock button battery asus_acpi ac lp snd_intel8x0 snd_ac97_codec ac97_bus snd_seq_dummy snd_seq_oss snd_seq_midi_event snd_seq snd_seq_device sg snd_pcm_oss floppy snd_mixer_oss snd_pcm cfi_cmdset_0002 cfi_util mtdpart snd_timer jedec_probe cfi_probe gen_probe snd ck804xrom sata_nv mtdcore chipreg i2c_nforce2 soundcore map_funcs libata snd_page_alloc pcspkr i2c_core k8temp hwmon forcedeth ohci1394 ieee1394 parport_pc ide_cd parport cdrom serio_raw scsi_transport_fc shpchp megaraid_mbox sd_mod scsi_mod megaraid_mm ext3 jbd ehci_hcd ohci_hcd uhci_hcd

Quoted 8 lines
May 15 21:10:31 vault1 kernel: Pid: 4256, comm: nfsd Not tainted 2.6.21.1 #1
May 15 21:10:31 vault1 kernel: RIP: 0010:[<ffffffff8025a710>] 
[<ffffffff8025a710>] cache_alloc_refill+0xe6/0x1f3
May 15 21:10:31 vault1 kernel: RSP: 0018:ffff81021dead6d0  EFLAGS: 00010002
May 15 21:10:31 vault1 kernel: RAX: 0000000000000001 RBX: 
ffff81012211d960 RCX: ffff810120013040
May 15 21:10:31 vault1 kernel: RDX: 000000000000000e RSI:

ffff810139020000 RDI: ffff810120013040

There wasn't a backtrace?

--b.

-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
_______________________________________________
NFS maillist - [email protected]
https://lists.sourceforge.net/lists/listinfo/nfs

author
#2 lore raw
Quoted 8 lines · J. Bruce Fields
Quoted 4 lines · Roger Heflin

Running bonnie over nfs on a RHEL4.4 client against a 2.6.21.1 server got me this crash after about 4 hours of running on the server:

This was running lvm -> ext3 -> nfs ---- nfsclient (RHEL4.4).

Yipes. Has this happened only once, or do you have a way to reliably reproduce it?

I have not reproduced it yet, I will update if I do, I suspect that I will be able to reproduce it, but it took several hours of running.

Quoted 2 lines

Is it a new problem? (And, if so, what changed?)

New different tests.

Quoted 29 lines
Quoted 3 lines
May 15 21:10:31 vault1 kernel: ------------[ cut here ]------------
May 15 21:10:31 vault1 kernel: kernel BUG at mm/slab.c:2380!

That's the check_spinlock_acquired() in cache_alloc_refill(). What causes that to fail?

Quoted 13 lines
May 15 21:10:31 vault1 kernel: invalid opcode: 0000 [1] SMP
May 15 21:10:31 vault1 kernel: CPU 0
May 15 21:10:31 vault1 kernel: Modules linked in: qla2xxx nfsd exportfs

lockd nfs_acl sunrpc hidp l2cap bluetooth ipv6 cpufreq_ondemand jfs dm_mirror dm_multipath dm_mod video sbs i2c_ec dock button battery asus_acpi ac lp snd_intel8x0 snd_ac97_codec ac97_bus snd_seq_dummy snd_seq_oss snd_seq_midi_event snd_seq snd_seq_device sg snd_pcm_oss floppy snd_mixer_oss snd_pcm cfi_cmdset_0002 cfi_util mtdpart snd_timer jedec_probe cfi_probe gen_probe snd ck804xrom sata_nv mtdcore chipreg i2c_nforce2 soundcore map_funcs libata snd_page_alloc pcspkr i2c_core k8temp hwmon forcedeth ohci1394 ieee1394 parport_pc ide_cd parport cdrom serio_raw scsi_transport_fc shpchp megaraid_mbox sd_mod scsi_mod megaraid_mm ext3 jbd ehci_hcd ohci_hcd uhci_hcd

Quoted 8 lines
May 15 21:10:31 vault1 kernel: Pid: 4256, comm: nfsd Not tainted 2.6.21.1 #1
May 15 21:10:31 vault1 kernel: RIP: 0010:[<ffffffff8025a710>] 
[<ffffffff8025a710>] cache_alloc_refill+0xe6/0x1f3
May 15 21:10:31 vault1 kernel: RSP: 0018:ffff81021dead6d0  EFLAGS: 00010002
May 15 21:10:31 vault1 kernel: RAX: 0000000000000001 RBX: 
ffff81012211d960 RCX: ffff810120013040
May 15 21:10:31 vault1 kernel: RDX: 000000000000000e RSI:

ffff810139020000 RDI: ffff810120013040

Quoted 4 lines

There wasn't a backtrace?

--b.

That was all I had before it took the machine out, it is on a serial console at this time, so if it happens again I should get better data.

Roger

-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
_______________________________________________
NFS maillist - [email protected]
https://lists.sourceforge.net/lists/listinfo/nfs